Dify

You are a professional customer service assistant responsible for:
- Answering user questions
- Providing product information
- Handling after-sales service
Please maintain a friendly and professional attitude.

Application Type: Chat Assistant
Model: gpt-4
System Prompt: |
  You are a professional AI assistant with the following capabilities:
  - Answering various questions
  - Assisting in problem-solving
  - Providing advice and guidance

  Please always maintain a friendly, accurate, and helpful attitude.
Temperature: 0.7
Max Length: 2000

Application Type: Workflow
Input: Document Upload
Processing Flow:
  1. Document Parsing
  2. Content Extraction
  3. Structured Analysis
  4. Generate Summary
Output: Analysis Report

Application Type: Chat Assistant
Model: gpt-4
System Prompt: |
  You are a professional programming assistant specializing in:
  - Code writing and optimization
  - Error debugging
  - Architecture design
  - Best practice recommendations

  Please provide clear and practical code solutions.

import requests

url = "https://your-dify-instance/v1/chat-messages"
headers = {
    "Authorization": "Bearer YOUR_APP_API_KEY",
    "Content-Type": "application/json"
}

data = {
    "inputs": {},
    "query": "Hello, please introduce yourself",
    "response_mode": "streaming",
    "user": "user_123"
}

response = requests.post(url, headers=headers, json=data)

# Multimodal input example
{
    "inputs": {
        "image": "data:image/jpeg;base64,...",
        "text": "Analyze the content in this image"
    },
    "query": "Please describe the image content in detail and provide analysis"
}

Development Environment:
  Model: gpt-3.5-turbo
  Max Length: 1000
  Temperature: 0.7

Production Environment:
  Model: gpt-4
  Max Length: 2000
  Temperature: 0.5

# Structured Prompt
## Role Definition
You are a professional [specific role]

## Task Description
Please help users with [specific task]

## Output Format
Please output in the following format:
1. Overview
2. Detailed Analysis
3. Recommendations

## Constraints
- Answers must be accurate
- Language should be clear
- Keep length within 500 words

Cache Settings:
  Enabled: true
  Expiration: 3600 seconds
  Cache Condition: Same Input

Concurrency Control:
  Max Concurrency: 10
  Queue Size: 100
  Timeout: 30 seconds

Resource Limits:
  Memory Limit: 2GB
  CPU Limit: 80%

# docker-compose.yml
version: '3.8'
services:
  dify-api:
    image: langgenius/dify-api:latest
    environment:
      - SECRET_KEY=your-secret-key
      - DB_HOST=postgres
      - REDIS_HOST=redis
      - OPENAI_API_KEY=your-apiyi-key
      - OPENAI_API_BASE=https://api.apiyi.com/v1
    depends_on:
      - postgres
      - redis

  dify-web:
    image: langgenius/dify-web:latest
    ports:
      - "3000:3000"
    depends_on:
      - dify-api

  postgres:
    image: postgres:14
    environment:
      - POSTGRES_DB=dify
      - POSTGRES_USER=dify
      - POSTGRES_PASSWORD=password

  redis:
    image: redis:alpine

# Monitoring script example
import requests
import time

def monitor_dify_health():
    try:
        response = requests.get("http://your-dify-instance/health")
        if response.status_code == 200:
            print("Dify running normally")
        else:
            print(f"Dify abnormal, status code: {response.status_code}")
    except Exception as e:
        print(f"Monitoring failed: {e}")

# Check every minute
while True:
    monitor_dify_health()
    time.sleep(60)